def plot(model, pos=None, scale=1, figsize=(15,8), verbose=3):
"""Plot the learned stucture.
Parameters
----------
model : dict
Learned model from the .fit() function..
pos : graph, optional
Coordinates of the network. If there are provided, the same structure will be used to plot the network.. The default is None.
scale : int, optional
Scaling parameter for the network. A larger number will linearily increase the network.. The default is 1.
figsize : tuple, optional
Figure size. The default is (15,8).
verbose : int, optional
Print progress to screen. The default is 3.
0: None, 1: ERROR, 2: WARN, 3: INFO (default), 4: DEBUG, 5: TRACE
Returns
-------
dict containing pos and G
pos : list
Positions of the nodes.
G : Graph
Graph model
"""
out = {}
G = nx.DiGraph() # Directed graph
layout='fruchterman_reingold'
# Extract model if in dict
if 'dict' in str(type(model)):
model = model.get('model', None)
# Bayesian model
if 'BayesianModel' in str(type(model)) or 'pgmpy' in str(type(model)):
if verbose>=3: print('[BNLEARN][plot] Making plot based on BayesianModel')
# positions for all nodes
pos = network.graphlayout(model, pos=pos, scale=scale, layout=layout)
# Add directed edge with weigth
# edges=model.edges()
edges=[*model.edges()]
for i in range(len(edges)):
G.add_edge(edges[i][0], edges[i][1], weight=1, color='k')
elif 'networkx' in str(type(model)):
if verbose>=3: print('[BNLEARN][plot] Making plot based on networkx model')
G = model
pos = network.graphlayout(G, pos=pos, scale=scale, layout=layout)
else:
if verbose>=3: print('[BNLEARN][plot] Making plot based on adjacency matrix')
G = network.adjmat2graph(model)
# Convert adjmat to source target
# df_edges=model.stack().reset_index()
# df_edges.columns=['source', 'target', 'weight']
# df_edges['weight']=df_edges['weight'].astype(float)
# # Add directed edge with weigth
# for i in range(df_edges.shape[0]):
# if df_edges['weight'].iloc[i]!=0:
# color='k' if df_edges['weight'].iloc[i]>0 else 'r'
# G.add_edge(df_edges['source'].iloc[i], df_edges['target'].iloc[i], weight=np.abs(df_edges['weight'].iloc[i]), color=color)
# Get positions
pos = network.graphlayout(G, pos=pos, scale=scale, layout=layout)
# Bootup figure
plt.figure(figsize=figsize)
# nodes
nx.draw_networkx_nodes(G, pos, node_size=500, with_labels=True, alpha=0.85)
# edges
colors = [G[u][v].get('color','k') for u,v in G.edges()]
weights = [G[u][v].get('weight',1) for u,v in G.edges()]
nx.draw_networkx_edges(G, pos, arrowstyle='->', edge_color=colors, width=weights)
# Labels
nx.draw_networkx_labels(G, pos, font_size=20, font_family='sans-serif')
# Get labels of weights
# labels = nx.get_edge_attributes(G,'weight')
# Plot weights
nx.draw_networkx_edge_labels(G, pos, edge_labels=nx.get_edge_attributes(G,'weight'))
# Making figure nice
ax = plt.gca()
ax.set_axis_off()
plt.show()
# Store
out['pos']=pos
out['G']=G
return(out)
def plot(model, pos=None, scale=1, width=15, height=8, verbose=3):
out = dict()
G = nx.DiGraph() # Directed graph
layout = 'fruchterman_reingold'
# Extract model if in dict
if 'dict' in str(type(model)):
model = model.get('model', None)
# Bayesian model
if 'BayesianModel' in str(type(model)) or 'pgmpy' in str(type(model)):
if verbose >= 3:
print('[BNLEARN.plot] Making plot based on BayesianModel')
# positions for all nodes
pos = network.graphlayout(model, pos=pos, scale=scale, layout=layout)
# Add directed edge with weigth
# edges=model.edges()
edges = [*model.edges()]
for i in range(len(edges)):
G.add_edge(edges[i][0], edges[i][1], weight=1, color='k')
elif 'networkx' in str(type(model)):
if verbose >= 3:
print('[BNLEARN.plot] Making plot based on networkx model')
G = model
pos = network.graphlayout(G, pos=pos, scale=scale, layout=layout)
else:
if verbose >= 3:
print('[BNLEARN.plot] Making plot based on adjacency matrix')
G = network.adjmat2graph(model)
# Convert adjmat to source target
# df_edges=model.stack().reset_index()
# df_edges.columns=['source', 'target', 'weight']
# df_edges['weight']=df_edges['weight'].astype(float)
#
# # Add directed edge with weigth
# for i in range(df_edges.shape[0]):
# if df_edges['weight'].iloc[i]!=0:
# color='k' if df_edges['weight'].iloc[i]>0 else 'r'
# G.add_edge(df_edges['source'].iloc[i], df_edges['target'].iloc[i], weight=np.abs(df_edges['weight'].iloc[i]), color=color)
# Get positions
pos = network.graphlayout(G, pos=pos, scale=scale, layout=layout)
# Bootup figure
plt.figure(figsize=(width, height))
# nodes
nx.draw_networkx_nodes(G, pos, node_size=500, with_labels=True, alpha=0.85)
# edges
colors = [G[u][v].get('color', 'k') for u, v in G.edges()]
weights = [G[u][v].get('weight', 1) for u, v in G.edges()]
nx.draw_networkx_edges(G,
pos,
arrowstyle='->',
edge_color=colors,
width=weights)
# Labels
nx.draw_networkx_labels(G, pos, font_size=20, font_family='sans-serif')
# Get labels of weights
#labels = nx.get_edge_attributes(G,'weight')
# Plot weights
nx.draw_networkx_edge_labels(G,
pos,
edge_labels=nx.get_edge_attributes(
G, 'weight'))
# Making figure nice
ax = plt.gca()
ax.set_axis_off()
plt.show()
# Store
out['pos'] = pos
out['G'] = G
return (out)